用大模型生成通用规划的抽象特征，自动纠错提升效果
Abstraction Generation for Generalized Planning with Pretrained Large Language Models
- 让大模型根据任务自动生成抽象特征和初始状态等要素
- 经自动调试后，部分大模型可生成有效抽象方案
- 适合研究通用规划与大模型结合的学者参考
定性数值规划（QNP）是通用规划（GP）的重要抽象模型，旨在一次性求解多个实例。近期研究表明大语言模型（LLMs）可作为通用规划器。本文探究了LLMs能否作为QNP抽象生成器，并设计自动化调试方法检测抽象错误，引导模型修正。提出一种提示协议：输入GP领域和训练任务，让LLM生成抽象特征，并将初始状态、动作集和目标抽象为QNP问题。实验表明，在自动化调试引导下，部分LLM能生成有效的QNP抽象。
原文摘要 · Abstract (English)
Qualitative Numerical Planning (QNP) serves as an important abstraction model for generalized planning (GP), which aims to compute general plans that solve multiple instances at once. Recent works show that large language models (LLMs) can function as generalized planners. This work investigates whether LLMs can serve as QNP abstraction generators for GP problems and how to fix abstractions via automated debugging. We propose a prompt protocol: input a GP domain and training tasks to LLMs, prompting them to generate abstract features and further abstract the initial state, action set, and goal into QNP problems. An automated debugging method is designed to detect abstraction errors, guiding LLMs to fix abstractions. Experiments demonstrate that under properly guided by automated debugging, some LLMs can generate useful QNP abstractions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。